Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Morphological divergence of the threatened Rocky Mountain sculpin (Cottus sp.) is driven by biogeography and flow regime: Implications for mitigating altered flow regime to freshwater fishes

Morphological divergence of the threatened Rocky Mountain sculpin (Cottus sp.) is driven by... INTRODUCTIONAlteration of stream hydrology by climate and human sources is predicted to have a major impact on freshwater fishes (Dudgeon et al., ). Stream hydrology is important for the consistency and maintenance of fish habitat, especially for resident or sedentary species, as they are unable to move away from suddenly changing environments. Alterations to stream hydrology, if severe enough, can reduce biodiversity (Pringle, Freeman, & Freeman, ). Determining how stream hydrology affects freshwater fishes is important for determining how to mitigate the impacts of stream hydrology and provide appropriate conservation measures.It is well known that local adaptations to various environments can lead to morphological and genetic differences within the same species (Collin & Fumagalli, ; Endler, ; Kawecki & Ebert, ). Even among similar flow gradients, morphological divergence can occur (McLaughlin & Grant, ; Pakkasmaa & Piironen, ). In freshwater systems, flow has been recognized as a driving force behind intraspecific morphological adaptation in fish species (Brinsmead & Fox, ; Langerhans, ; Langerhans, Layman, Langerhans, & Dewitt, ). In these studies, pelagic fish tend to have more slender, fusiform bodies and deeper caudal peduncles when exposed to faster‐flowing waters (Collin & Fumagalli, ; McLaughlin & Grant, ; Webb, ). These morphological adaptations help minimize drag forces on the body, thereby optimizing energetic expenditure (Sagnes & Statzner, ; Webb, ). Few studies account for benthic fish, however, which have different life histories (Facey & Grossman, ).The Rocky Mountain sculpin (Cottus sp.) is a ‘Threatened’ species in Canada and is listed under the federal Species at Risk Act. Like many legal frameworks around the globe (Vardakas et al., ), the Species at Risk Act provides protection of ‘critical habitat’, defined as ‘the habitat that is necessary for the survival or recovery of a listed wildlife species’ (Fisheries and Oceans Canada, ). In addition, a recovery plan that outlines the actions needed for species recovery is required. This plan must address how to mitigate key threats to the species. For Rocky Mountain sculpin, the main threat is thought to be changes in stream hydrology. Rocky Mountain sculpin live in varying hydrological regimes, differing by three orders of magnitude across its range (from minimum 0–1 m3 s‐1 to >120 m3 s‐1; Figure ). These differences in stream hydrology are due in large part to stream alterations to improve irrigation for agriculture. For example, the St. Mary's River was diverted in northern Montana to the Milk River basin to improve irrigation. In recent years, this has changed the flow in the North Milk River from about 7.5 m3 s‐1 during natural conditions to upwards of 17m3 s‐1 during augmentation. Given the alteration of the natural hydrology of systems occupied by Rocky Mountain sculpin and the difference in magnitude of hydrological regimes, Rocky Mountain sculpin provide a model species for understanding impacts of stream hydrology on differences in population levels for freshwater fishes.1Map of study locations (top) and associated hydrographs (bottom). Fish samples of populations of Rocky Mountain sculpin were collected from the Flathead River (blue), Lee Creek (orange), St. Mary River (red), and the North Milk River (green). Hydrographs were developed from 84 to 104 years of data collected by four representative gauging stations (Water Survey of Canada, )Bottom‐dwelling fish exhibit morphological differences for a variety of reasons, including habitat characteristics such as substrate (Whiteley, Gende, Gharrett, & Tallmon, ), thermal conditions (Koumoundouros et al., ), and flow regime (Natsumeda, Tsuruta, Takeshima, Awata, & Iguchi, ). Sculpins are sedentary fish with relatively deep bodies and a bulky caudal peduncle (Webb, ). Their robust torso allows for short, strong bursts of forward movement when necessary, but prolonged swimming is energetically costly. Sculpins maintain their position in running water by relying heavily on their large, rigid pectoral fins to hold themselves in place among the substrate (Facey & Grossman, ). In highly turbulent environments, sculpin exhibit prominent pectoral fins forming more robust tissue connections where the fin meets the body (Kane & Higham, ).Unlike pelagic species, benthic species must morphologically reduce drag against their body in river systems to maintain their position (Koehl, ). Because Rocky Mountain sculpin have survived in varying hydrological regimes, Rocky Mountain sculpin should have a more streamlined body shape to reduce drag and/or stronger pectoral fins to grip onto substrate in the higher flow systems. Higher velocities may also select for differences in pore and fin ray counts to account for different demands on flow compensation and prey detection rates. This study compared the body morphology and fin ray and pore counts of Rocky Mountain sculpin across four populations in Canada which vary in stream hydrology and biogeography. Specifically, the objectives were to understand: (1) if Rocky Mountain sculpin exhibit morphological adaptations to stream flow across their four prominent river systems in Alberta and British Columbia, Canada, and (2) whether there is a difference in meristic counts between Rocky Mountain sculpin across populations. This information has built upon the known ecology of this species and will help in identifying whether populations from different river basins and biogeography are unique in their meristics and morphometrics.METHODSSpeciesThe identity and description of sculpins (family Cottidae), including Rocky Mountain sculpin, have long confused researchers in western North America (Bajkov, ; Hughes & Peden, ; Lemoine et al., ; Markle & Hill, ; Taylor, ; Zimmerman & Wooten, ). Previously, Rocky Mountain sculpin were confused with at least three other species (COSEWIC, ). Rocky Mountain sculpin were first characterized in 2002 (Neely, ), and are considered a taxonomically valid species in Canada (Taylor, ), despite no official description. Rocky Mountain Sculpin show clear phylogenetic differences with other Cottus species (COSEWIC, ; Neely, ; Young, McKelvey, Pilgram, & Schwartz, ). Differences in population levels do occur between populations on the eastslope and westslope of the Rocky Mountains in Canada (Ruppert et al., ), but analysis of 1140 cytochrome b base pairs show that these populations are the same species (COSEWIC, ). Thus, despite historical confusion over species identity, contemporary data clearly show Rocky Mountain sculpin, from both eastslope and westslope of the Rocky Mountains, as the same, unique species.Study areaRocky Mountain sculpin occur across a wide range of stream hydrology (Figure ). Of the four study systems, the Flathead River has the highest flow, with average peak discharge rate of 125 m3 s‐1. The St. Mary River has the highest average peak flow rate for streams on the eastern slopes of the Rocky Mountains with a mean of 61 m3 s‐1. The next highest flow rate occurs in Lee Creek with average peak flows at 9 m3 s‐1. North Milk River is the slowest river with mean peak flow rate of 8 m3 s‐1. Much of this flow is through augmentation from the St. Mary's River. Built in 1917, the St. Mary canal was designed to divert water from the St. Mary River to the Milk River basin during the growing season to supplement flows for irrigation of crops in the Milk River basin.Sculpin collectionA minimum of 40 Rocky Mountain sculpins were collected from each of the four rivers. Sculpins were electro‐shocked using a Smith‐Root LR‐24 backpack electrofisher. Sculpins were left overnight in a flow‐through bin to reduce their stress levels before transport. In total, 339 live fish were moved to holding tanks in the Aquatics Research Facility at the University of Alberta, a level‐three bio‐secure aquatic holding facility.Sculpin data preparationRocky Mountain sculpin collected from each river were euthanized and positioned with splayed fins. Images were taken with a Nikon D3100 digital single‐lens reflex camera equipped with a Nikon DX AF‐S Nikkor 18–55 mm zoom lens set at 35 mm, 225 ISO, 5.3 f‐stop, and 1/60 shutter speed. Torsos were placed at a level plane with the camera lens from head to caudal fin to prevent a warped, disproportionate representation of shape in the digital photos. Digital photos of the dorsal and lateral perspectives of the fish were taken in RAW (.NREF) form. Each photograph included a reference scaling factor to standardize fish size across photographs. Meristic measurements, such as fin rays, fin spines, and head/body pores, were subsequently determined for each individual. These meristic counts were chosen as they were similar to the original species description (2002).Landmarks were placed in areas that provided easy replication, such as the location of the eye and the origin and insertion of fin locations. Landmarks were marked using tpsDIG software (Rohlf, ), and their location was translated into X and Y coordinates in a .TPS file. Landmark criteria, as described by Dryden and Mardia (), were reference points on the sculpin's body that could be found accurately and marked across a large number of specimens. These points included fin insertion points, eye placement, and caudal peduncle locations.Differences in body morphologyDifferences in body morphology were tested using geometric morphometric analysis using the geomorph in the R programming language (Adams, Collyer, & Sherratt, ; R Core Team, ). A General Procrustes Analysis (GPA) was conducted on the coordinates. The GPA optimally superimposed landmarks by rotating, sizing and centralizing them without compromising the overall shape from the coordinates (Rohlf, ; Rohlf & Marcus, ; Slice, ). This process produced useable X‐Y Procrustes residual coordinates that could be used for a variety of multivariate analyses. In addition, the difference between Euclidean distances between the Procrustes landmarks was measured and compared for each population. Euclidean Distance Matrix Analysis (EDMA) was used to identify variance across the mean landmark positions between groups (Lele, ).To determine if there were differences in overall shape across populations, a distance‐based Procrustes Analysis of Variance (ANOVA) using 1000 permutations was used to test between GPA residuals of dorsal and lateral landmarks. The Procrustes ANOVA allowed a simultaneous analysis of sum‐of‐squared Procrustes distances across all the coordinates instead of considering one coordinate at a time (Goodall, ). This allowed the determination of significant differences in overall shape across populations using permutation analyses to generate reliable P‐values (Goodall, ; Sherratt, ). Pairwise comparisons were conducted using permutational t‐tests, with 1000 resampling iterations to determine which populations were significantly different using pairwiseD.test in geomorph. This test is specialized for comparing geometric shape variation across groups by using the Procrustes residual values that represent shape variation, and calculating P‐values based on the Euclidean distances between the mean of each population (Adams et al., ; Collyer, Sekora, & Adams, ). To visualize the results of the pairwise comparisons, coordinate data for each sculpin were converted into a single warp score using candisc (Friendly & Fox, ). Warp scores were generated based on the degree and location of morphometric variance relative to the other fish in the study (Webster & Sheets, ; Zelditch, Swiderski, & Sheets, ). Means of the values in each population were calculated as well as the Euclidean distance between those means (Sherratt, ). After 1000 permutations, a Canonical Variates Analysis (CVA) was used to determine differences between groups that were calculated in the pairwise comparison. As a form of discriminatory analysis, CVA was used because it can measure differences in overall shape between the four predefined vectors (Rohlf, ; Zelditch et al., ). The vectors identified landmarks that were major drivers in population variation. This generated the data necessary for visualization and the vegan package (Oksanen et al., ) was used to help customize the visualization.Meristic differences in body formFin ray and pore counts were compared across populations using a permutational ANOVA. Permutational ANOVA is useful when datasets do not adhere to the assumptions of a traditional ANOVA (i.e. independence, normality, and homogeneity of variances), which is common in natural and biological realms (Anderson, ). To test significance across groups, a pairwise Wilcoxon Rank Sum Test was used and a Holm P‐value adjustment was applied to correct for Type I errors when multiple comparisons are made (Aickin & Gensler, ).RESULTSDifferences in body morphologyThere was strong divergence in body morphology of Rocky Mountain sculpin across its Canadian range. The Flathead River population was the most visibly divergent population among the four populations of Rocky Mountain sculpin. From the dorsal perspective, the caudal peduncle and snout positions relative to pectoral fin insertion points were the primary features that differentiated the westslope population from the eastslope groups (P<0.001; Figures  and ). In addition, westslope Rocky Mountain sculpins had stunted heads, creating a more antero‐posteriorly compressed head shape, and the length of the torso beyond the pectoral fin insertion points was significantly narrower and longer. From the lateral perspective, the westslope population was significantly different from the eastslope populations (P<0.001; Table ). Eye placement was oriented closer to the insertion point of the dorsal spines, contributing to a more dorsally flattened form (Figure ). The placement of the isthmus showed significant differentiation in the Flathead River population (Figure , P < 0.01). It was located closer to the snout, probably as the result of a more antero‐posteriorly compressed head, as seen in the dorsal view. The westslope population also had the longest, narrowest, and flattest body shape of the four groups.2Differences in dorsal Procrustes scores of Rocky Mountain sculpin (Cottus sp.) grouped by river system. Using Canonical Variate analysis, the distance between centroids (crosshairs) of populations was maximized. Populations included the Flathead River (blue), Lee Creek (orange), St. Mary River (red), and North Milk River (green)3Differences in lateral Procrustes scores of Rocky Mountain sculpin (Cottus sp.) grouped by river system. Using Canonical Variate analysis, the distance between centroids (crosshairs) of populations was maximized. Populations included the Flathead River (blue), Lee Creek (orange), St. Mary River (red), and North Milk River (green)4Dorsal (top) and lateral (bottom) landmarks of Rocky Mountain sculpin (Cottus sp.) across its Canadian range, including the Flathead River (blue), Lee Creek (orange), St. Mary River (red), and North Milk River (green). Dorsal landmark placements were amplified 4.5 times to show the differences across populations. Lateral landmark placements had to be amplified by only three times to show differences visually. The X and Y axes represent the vector scores once standardized using General Procrustes AnalysisPairwise differences in morphometric landmarks of Rocky Mountain sculpin (Cottus sp.) across populations including the Flathead River, Lee Creek, North Milk River, and St. Mary River. Shown are permutated t‐tests using Euclidean distances between the means of dorsal (top corner) and lateral (bottom corner) landmarks across each river. P‐values are based on 1000 permutations and are adjusted using a Holm correction. Asterisks indicate significant differencesFlatheadLee CreekNorth MilkSt. MaryFlathead‐0.001*0.001*0.001*Lee Creek0.001*‐0.001*0.13North Milk0.001*0.001*‐0.001*St. Mary0.001*0.001*0.001*‐Differences in body morphology were strongly divergent not just across the continental divide, but also within the eastslope populations. All of the three eastslope populations showed a significant difference in shape (P <0.001; Table ). The differences in the North Milk River population were pronounced in the pectoral fin insertion point locations (Figure ). The width between the anterior insertion points in the dorsal view showed that the North Milk River population had the widest overall shape (Figure ). The close proximity of the posterior and anterior pectoral fin insertion points led to the North Milk River population having the smallest fin base (Figure ). Lee Creek sculpins had lower and posteriorly placed pelvic fin insertion points, leading to the most dorso‐ventrally broad shape of the four populations (Figure ). St. Mary River sculpin eye placement was closest to the snout and the dorsal spine insertion point is almost level with the eyes, contributing to the most dorso‐ventrally flattened eastslope population (Figure ).Meristic differences in body formThere were significant meristic differences in body form of Rocky Mountain sculpin across its Canadian range. The average pore counts and fin ray/spine counts among populations were significantly different (permANOVA, P<0.001). Pore counts varied across populations by no more than five pores and fin ray/spine counts varied only by one in populations that were significantly different (Table ).Summary of meristic fin and pore counts of Rocky Mountain sculpin (Cottus sp.) across its Canadian range, including the Flathead River, Lee Creek, the North Milk River, and the St. Mary River. Shown are means, ranges, and standard deviationsRiver PopulationsFlatheadLee CreekNorth MilkSt. MaryCharacterMeanRangeSDMeanRangeSDMeanRangeSDMeanRangeSDPoresInfraorbital106‐10±1.5108‐10±0.3108‐14±1.1106‐10±0.5Pre‐operculo‐mandibular1410‐16±1.31411‐18±0.91412‐18±1.11412‐16±0.6Mandibular1310‐16±1.41712‐18±1.5157‐20±2.91710‐18±1.6Posterior orbital1712‐22±2.11611‐18±1.71711‐22±2.41710‐22±2.6Forehead22±011‐3±0.421‐2±0.511‐2±0.4Lateral line2419‐32±2.72415‐30±3.12611‐34±4.22314‐36±3.0Fin ray/spine countsDorsal spines86‐9±0.787‐9±0.687‐9±0.687‐8±0.5Dorsal rays1713‐19±1.21716‐19±0.91715‐20±0.81713‐19±1.3Anal rays1211‐14±0.81211‐13±0.71211‐15±0.71210‐13±0.8Pectoral rays1311‐14±0.61210‐13±1.1129‐13±1.0129‐14±1.0Pelvic rays44±044±044±044±0Caudal rays1312‐14±0.51210‐13±0.8129‐13±0.91211‐15±1.0Opercular spines11±011±011±011±0N4174106118The westslope population exhibited several significant meristic differences from the eastslope populations. They had significantly more forehead pores (P<0.001), and fewer mandibular pores (P<0.001; Tables  and ). Furthermore, the westslope population had more pectoral (P<0.001) and caudal fin rays (P<0.001) than the Lee Creek and North Milk River population (Tables  and ).Pairwise differences in pore counts of Rocky Mountain sculpin (Cottus sp.) across Canada, including the Flathead River, Lee Creek, the North Milk River, and the St. Mary River. Pairwise Wilcoxon Rank Sum Test P‐values are displayed with Holm corrected significance levels. Asterisks indicate significant resultsInfraorbital poreMandibular poreForehead poreLateral line poreFlathead : Lee Creek0.91<0.001*<0.001*0.67Flathead : North Milk1.00<0.001*<0.001*0.02Flathead : St. Mary0.81<0.001*<0.001*0.06Lee Creek : North Milk0.220.002*<0.001*0.01Lee Creek : St. Mary1.000.481.000.21North Milk : St. Mary0.080.002*<0.001*<0.001*Pairwise differences in fin ray and spine counts of Rocky Mountain sculpin (Cottus sp.) across Canada, including the Flathead River, Lee Creek, the North Milk River, and the St. Mary River. Pairwise Wilcoxon Rank Sum Test P‐values are displayed with Holm corrected significance levels. Asterisks indicate significant resultsDorsal fin spinesDorsal fin raysAnal fin raysPectoral fin raysCaudal fin raysFlathead : Lee Creek0.640.991.00<0.001*<0.001**Flathead : North Milk0.071.001.00<0.001*<0.001**Flathead : St. Mary0.051.001.00<0.001*0.02Lee Creek : North Milk0.730.991.001.000.81Lee Creek : St. Mary0.731.001.001.000.81North Milk : St. Mary0.731.001.000.710.81There were some meristic variations among the eastslope populations. Lee Creek and North Milk River populations differed in three of the pore counts, where the Lee Creek population had significantly more mandibular pores (P=0.002), but fewer forehead (P<0.001) pores (Tables ). Similar to Lee Creek, the St. Mary River population had a higher number of mandibular pores (P=0.002) than the North Milk River population, as well as fewer forehead (P<0.001) and lateral line pores (P<0.001; Tables  and ). There were no differences in fin ray/spine counts across east slope populations. There were no meristic differences between the Lee Creek and St. Mary populations.DISCUSSIONMany species exhibit morphological divergence in relation to environmental gradients (Endler, ; Kawecki & Ebert, ). Rocky Mountain sculpin exhibited strong morphological and meristic body differences across broad hydrological gradients. Specifically, higher flow rates were correlated with higher adaptive morphological divergence. For example, populations exhibited strong divergence in meristics across the hydrological gradient. The population with highest flow (westslope) had more caudal fin rays, which help provide thrust, and pectoral fin rays to increase friction against substrate in high flow environments (Kane & Higham, ; Taft, Lauder, & Madden, ; Webb, ). The highest flow population also had fewer mandibular pores and more forehead pores than the lower flow (eastslope) population. In lotic systems, sculpin use their head pores to detect suspended prey and mandibular pores to extract buried prey (Hoekstra & Janssen, ). The increase in forehead pores helps to detect drifting prey in fast currents. The reduction in mandibular pores is an example of a feature that cannot be justified by flow regime, and is possibly the result of divergence based on prolonged biogeographic isolation from other populations.Biogeographic distance can play an important role in determining morphological divergence. The continental divide has undoubtedly influenced body morphology between westslope and eastslope Rocky Mountain sculpin populations given that the populations have probably been isolated for about 10 000 years (Nelson & Paetz, ). The westslope populations exhibited a more compressed and elongated torso than the eastslope populations, probably because of the interplay between prolonged biogeographic isolation and different flow regimes (Langerhans et al., ). If flow regime were the primary cause of variation, there would have been more similar morphologies between the St. Mary and Flathead River populations, and almost identical morphologies between the Lee Creek and North Milk River populations. Instead, there were closer morphometric and meristic values between the Lee Creek and St. Mary River populations (the closest geographically), as Lee Creek is a tributary of the St. Mary River. This theory is also validated by the large degree of phenotypic variation between the westslope populations and the eastslope populations. Gene flow is a major influence on the success of local adaptation (Kawecki & Ebert, ), therefore the disconnection from the eastslope populations undoubtedly contributes to the observed morphological divergence.Phenotypic variation is required to optimize fitness, especially in relation to stochastic environmental events (McGuigan, Franklin, Moritz, Blows, & Wainwright, ; Taylor & McPhail, ). Since phenotypic diversity parallels genetic diversity in Rocky Mountain sculpin (Ruppert et al., ), there is the possibility that under persistent selection Rocky Mountain sculpin will have maladaptive morphological features with changing stream flows. For example, under baseflow conditions, Rocky Mountain sculpin with narrower and shorter caudal peduncles were shown to have poorer swimming performance (Veillard, Ruppert, Tierney, Watkinson, & Poesch, ). With climate change expected to rapidly alter stream hydrology, these traits may be maladaptive, making it difficult for Rocky Mountain sculpin to adapt to rapidly changing conditions (Stearns & Kawecki, ).Understanding how biogeography and stream hydrology influences body morphology is important to improve species conservation. Rocky Mountain sculpin are a threatened species in Canada and could be further imperilled if their dominant phenotypes are incapable of adapting to altered river conditions such as flood events and drought conditions (Lytle & Poff, ). Their limited dispersal can lead to specialized morphologies and extreme events could eliminate a population. Management efforts should be directed, therefore, toward preserving genetic diversity at the population level of the species (Riffel & Schreiber, ; Ruppert et al., ), while focusing future research toward understanding how genotypic and phenotypic divergence change in relation to stream hydrology. Almost all fisheries management programmes are developed around preserving genetic variation, thereby protecting the species’ ability to overcome unpredicted environmental circumstances (Fraser & Bernatchez, ).CONCLUSIONMany studies have shown that life‐history characteristics are good predictors of extinction risk. These studies provide an important linkage to how biological features such as reproductive traits, habitat, and age/growth relationships can make species more susceptible to decline or put them more generally at risk (Glass, Corkum, & Mandrak, ; Purvis, Gittleman, Cowlishaw, & Mace, ; Stark, Banks, & Vargas, ). These relationships have intuitive appeal as it is easy to see the linkages between life‐history characteristics such as small body size, and understand how they may lead to limited dispersal and potentially a lack of connectivity between populations (and potential rescue). However, despite the importance of phenotypic plasticity in shaping local adaptations, the utility of phenotypic or morphological diversity has remained relatively understudied in the conservation literature. Here, a combination of biogeography and flow regime appear to be driving phenotypic and morphological divergence between Rocky Mountain sculpin populations. High‐flow lotic systems have shifted Rocky Mountain sculpin towards minimizing body depth. Results from this study suggest that small physical differences within the range of a fish species can have impacts on the energetic ability of the Rocky Mountain sculpin to exist in a wide range of lotic environments. By incorporating morphological diversity within the context of life‐history considerations already being used in assessing species, the potential of species to respond to both natural and human alterations can be predicted more accurately.ACKNOWLEDGEMENTSFunding for this project was provided by Fisheries and Oceans Canada, Species at Risk (SARCEP) (DW, MP), and NSERC Discovery Grant (MP). The authors thank Marie Veillard, Denyse Dawe, Christopher Smith, Elashia Young, Caitlin Good, Kenton Neufeld, Bryan Maitland, and Doug Watkinson for assisting with field collection and laboratory work. The care and assistance importing fish to the University of Alberta Aquatics Facility was provided by Toni Bayans and Simmone Kerswell. Additional thanks to Dr Jonathan Ruppert for his helpful advice on data analysis and Hedin Nelson‐Chorney for his critiques on early manuscript drafts. The procedures for this project was conducted under an approved Canadian Council of Animal Care Approval (AUP 00000759), Alberta Fish Research Licences 14‐2415AFR and 16‐0101FR, Alberta Fish Import Licence FIL2015‐0074, Federal Introductions and Transfers Licence VI16‐236322, and British Columbia Fish Collection Permit CB15‐171090.REFERENCESAdams, D., Collyer, M., & Sherratt, E. (2014). geomorph: Software for geometric morphometric analyses. R package version, 2.Aickin, M., & Gensler, H. (1996). Adjusting for multiple testing when reporting research results: The Bonferroni vs Holm methods. American Journal of Public Health, 86, 726–728.Anderson, M. J. (2001). Permutation tests for univariate or multivariate analysis of variance and regression. Canadian Journal of Fisheries and Aquatic Sciences, 58, 626–639.Bajkov, A. (1927). Reports of the Jasper Park Lakes Investigations, 1925‐26. I. The fishes. Contributions to Canadian Biology and Fisheries, 3, 379–404.Brinsmead, J., & Fox, M. (2002). Morphological variation between lake‐ and stream‐dwelling rock bass and pumpkinseed populations. Journal of Fish Biology, 61, 1619–1638.Collin, H., & Fumagalli, L. (2011). Evidence for morphological and adaptive genetic divergence between lake and stream habitats in European minnows (Phoxinus phoxinus, Cyprinidae). Molecular Ecology, 20, 4490–4502.Collyer, M., Sekora, D., & Adams, D. (2015). A method for analysis of phenotypic change for phenotypes described by high‐dimensional data. Heredity, 115, 357–365.COSEWIC. (2010). COSEWIC assessment and status report on the Rocky Mountain Sculpin Cottus sp., Westslope populations, in Canada. Committee on the Status of Endangered Wildlife in Canada. Ottawa.Dryden, I. L., & Mardia, K. V. (2016). Statistical shape analysis: With applications in R. Chichester, UK: John Wiley & Sons.Dudgeon, D., Arthington, A. H., Gessner, M. O., Kawabata, Z. I., Knowler, D. J., Lévêque, C., … Stiassny, M. L. (2006). Freshwater biodiversity: Importance, threats, status and conservation challenges. Biological Reviews, 81, 163–182.Endler, J. A. (1986). Natural selection in the wild. Princeton, NJ: Princeton University Press.Facey, D. E., & Grossman, G. D. (1990). The metabolic cost of maintaining position for four North American stream fishes: Effects of season and velocity. Physiological Zoology, 63, 757–776.Fisheries and Oceans Canada. (2017). Action Plan for the Milk River and St. Mary River Drainage Basins in Canada. Fisheries and Oceans, Species at Risk Act Action Plan Series, Ottawa, Canada.Fraser, D. J., & Bernatchez, L. (2001). Adaptive evolutionary conservation: Towards a unified concept for defining conservation units. Molecular Ecology, 10, 2741–2752.Friendly, M., & Fox, J. (2010). Candisc: R package for canonical discriminant analysis: Accessed at: http://cran.r‐project.org/web/packages/candisc.Glass, W. R., Corkum, L. D., & Mandrak, N. E. (2017). Living on the edge: Traits of freshwater fish species at risk in Canada. Aquatic Conservation: Marine and Freshwater Ecosystems, 27, 938–945.Goodall, C. (1991). Procrustes methods in the statistical analysis of shape. Journal of the Royal Statistical Society. Series B (Methodological), (1991), 285–339.Hoekstra, D., & Janssen, J. (1985). Non‐visual feeding behavior of the mottled sculpin, Cottus bairdi, in Lake Michigan. Environmental Biology of Fishes, 12, 111–117.Hughes, G. W., & Peden, A. E. (1984). Life history and status of the shorthead sculpin (Cottus confusus: Pisces, Cottidae) in Canada and the sympatric relationship to the slimy sculpin (Cottus cognatus). Canadian Journal of Zoology, 62, 306–311.Kane, E. A., & Higham, T. E. (2012). Life in the flow lane: Differences in pectoral fin morphology suggest transitions in station‐holding demand across species of marine sculpin. Zoology, 115, 223–232.Kawecki, T. J., & Ebert, D. (2004). Conceptual issues in local adaptation. Ecology Letters, 7, 1225–1241.Koehl, M. (1984). How do benthic organisms withstand moving water? American Zoologist, 24, 57–70.Koumoundouros, G., Ashton, C., Sfakianakis, D., Divanach, P., Kentouri, M., Anthwal, N., & Stickland, N. (2009). Thermally induced phenotypic plasticity of swimming performance in European sea bass Dicentrarchus labrax juveniles. Journal of Fish Biology, 74, 1309–1322.Langerhans, R. B. (2008). Predictability of phenotypic differentiation across flow regimes in fishes. Integrative and Comparative Biology, 48, 750–768.Langerhans, R. B., Layman, C. A., Langerhans, A. K., & Dewitt, T. J. (2003). Habitat‐associated morphological divergence in two Neotropical fish species. Biological Journal of the Linnean Society, 80, 689–698.Lele, S. (1993). Euclidean distance matrix analysis (EDMA): Estimation of mean form and mean form difference. Mathematical Geology, 25, 573–602.Lemoine, M., Young, M. K., McKelvey, K. S., Eby, L., Pilgram, K. L., & Schwartz, M. K. (2014). Cottus schitsuumsh, a new species of sculpin (Scorpaeniformes: Cottidae) in the Columbia River basin, Idaho‐Montana, USA. Zootaxa, 3755, 241–258.Lytle, D. A., & Poff, N. L. (2004). Adaptation to natural flow regimes. Trends in Ecology & Evolution, 19, 94–100.Markle, D. F., & Hill, D. L. (2000). Taxonomy and distribution of the Malheur mottled sculpin, Cottus bendirei. Northwest Science, 74, 202–211.McGuigan, K., Franklin, C. E., Moritz, C., Blows, M. W., & Wainwright, P. (2003). Adaptation of rainbow fish to lake and stream habitats. Evolution, 57, 104–118.McLaughlin, R. L., & Grant, J. W. (1994). Morphological and behavioural differences among recently‐emerged brook charr, Salvelinus fontinalis, foraging in slow‐ vs. fast‐running water. Environmental Biology of Fishes, 39, 289–300.Natsumeda, T., Tsuruta, T., Takeshima, H., Awata, S., & Iguchi, K. I. (2014). Variation in morphological characteristics of Japanese fluvial sculpin related to different environmental conditions in a single river system in eastern Japan. Ecology of Freshwater Fish, 23, 114–120.Neely, D. A. (2002). A systematic and taxonomic reassessment of the mottled sculpin species complex, Cottus bairdii Girard (Teleostei: Cottidae) (PhD thesis). University of Alabama, Tuscaloosa, AB.Nelson, J. S., & Paetz, M. J. (1992). The fishes of Alberta. Edmonton, Alberta: University of Alberta Press.Oksanen, J., Blanchet, F., Kindt, R., Legendre, P., Minchin, P., O’Hara, R., … Stevens, H. (2013). Vegan: Community Ecology Package. 2013. R‐package version 2.0‐10. Available from: https://cran.r‐project.org/package=vegan.Pakkasmaa, S., & Piironen, J. (2000). Water velocity shapes juvenile salmonids. Evolutionary Ecology, 14, 721–730.Pringle, C. M., Freeman, M. C., & Freeman, B. J. (2000). Regional effects of hydrologic alterations on riverine macrobiota in the New World: Tropical–temperate comparisons. BioScience, 50, 807–823.Purvis, A., Gittleman, J. L., Cowlishaw, G., & Mace, G. M. (2000). Predicting extinction risk in declining species. Proceedings of the Royal Society of London B: Biological Sciences, 267, 1947–1952.R Core Team (2016). R: A language and environment for statistical computing. Vienna, Austria: R Foundation for Statistical Computing.Riffel, M., & Schreiber, A. (1998). Morphometric differentiation in populations of the Central European sculpin Cottus gobio L., a fish with deeply divergent genetic lineages. Canadian Journal of Zoology, 76, 876–885.Rohlf, F. J. (1999). Shape statistics: Procrustes superimpositions and tangent spaces. Journal of Classification, 16, 197–223.Rohlf, F. J. (2005). tpsDig, digitize landmarks and outlines, version 2.05. Department of Ecology and Evolution, State University of New York at Stony Brook.Rohlf, F. J., & Marcus, L. F. (1993). A revolution in morphometrics. Trends in Ecology & Evolution, 8, 129–132.Ruppert, J. L., James, P. M., Taylor, E. B., Rudolfsen, T., Veillard, M., Davis, C. S., … Poesch, M. S. (2017). Riverscape genetic structure of a threatened and dispersal limited freshwater species, the Rocky Mountain Sculpin (Cottus sp.). Conservation Genetics, 2017, 1–13.Sagnes, P., & Statzner, B. (2009). Hydrodynamic abilities of riverine fish: A functional link between morphology and velocity use. Aquatic Living Resources, 22, 79–91.Sherratt, E. (2014). Quick guide to Geomorph v. 2.0. Available from: http://www.public.iastate.edu.Slice, D. E. (2001). Landmark coordinates aligned by Procrustes analysis do not lie in Kendall's shape space. Systematic Biology, 50, 141–149.Stark, J. D., Banks, J. E., & Vargas, R. (2004). How risky is risk assessment: The role that life history strategies play in susceptibility of species to stress. Proceedings of the National Academy of Sciences of the United States of America, 101, 732–736.Stearns, S. C., & Kawecki, T. J. (1994). Fitness sensitivity and the canalization of life‐history traits. Evolution, (48), 1438–1450.Taft, N., Lauder, G., & Madden, P. (2008). Functional regionalization of the pectoral fin of the benthic longhorn sculpin during station holding and swimming. Journal of Zoology, 276, 159–167.Taylor, E. B. (2010). Changes in taxonomy and species distributions and their influence on estimates of faunal homogenization and differentiation in freshwater fishes. Diversity and Distributions, 16, 676–689.Taylor, E. B., & McPhail, J. (1985). Variation in body morphology among British Columbia populations of coho salmon, Oncorhynchus kisutch. Canadian Journal of Fisheries and Aquatic Sciences, 42, 2020–2028.Vardakas, L., Kalogianni, E., Papadaki, C., Vavalidis, T., Mentzafou, A., & Koutsoubas, D. (2017). Defining critical habitat conditions for the conservation of three endemic and endangered cyprinids in a Mediterranean intermittent river before the onset of drought. Aquatic Conservation: Marine and Freshwater Ecosystems. https://doi.org/10.1002/aqc.2735.Veillard, M. F., Ruppert, J. L., Tierney, K., Watkinson, D. A., & Poesch, M. (2017). Comparative swimming and station‐holding ability of the threatened Rocky Mountain Sculpin (Cottus sp.) from four hydrologically distinct rivers. Conservation Physiology, 5, 1–12.Water Survey of Canada. (2016). Wateroffice: Historical Hydrometric Data: Environment Canada, Government of Canada.Webb, P. W. (1984). Form and function in fish swimming. Scientific American, 251, 58–68.Webster, M., & Sheets, H. D. (2010). A practical introduction to landmark‐based geometric morphometrics. Quantitative Methods in Paleobiology, 16, 168–188.Whiteley, A. R., Gende, S. M., Gharrett, A. J., & Tallmon, D. A. (2009). Background matching and color‐change plasticity in colonizing freshwater sculpin populations following rapid deglaciation. Evolution, 63, 1519–1529.Young, M. K., McKelvey, K. S., Pilgram, K. L., & Schwartz, M. K. (2013). DNA barcoding at riverscape scales: Assessing biodiversity among fishes of the genus Cottus (Teleostei) in northern Rocky Mountain streams. Molecular Ecology Resources, 13, 583–595.Zelditch, M. L., Swiderski, D. L., & Sheets, H. D. (2012). Geometric morphometrics for biologists: A primer (2nd ed.). San Diego, CA: Elsevier.Zimmerman, E. G., & Wooten, M. C. (1981). Allozymic variation and natural hybridization in sculpins, Cottus confusus and Cottus cognatus. Biochemical Systematics and Ecology, 9, 341–346. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Aquatic Conservation: Marine and Freshwater Ecosystems Wiley

Morphological divergence of the threatened Rocky Mountain sculpin (Cottus sp.) is driven by biogeography and flow regime: Implications for mitigating altered flow regime to freshwater fishes

Download PDF
9 pages

Loading next page...
 
/lp/wiley/morphological-divergence-of-the-threatened-rocky-mountain-sculpin-4h9sgN8vR8

References (62)

Publisher
Wiley
Copyright
Copyright © 2018 John Wiley & Sons, Ltd.
ISSN
1052-7613
eISSN
1099-0755
DOI
10.1002/aqc.2866
Publisher site
See Article on Publisher Site

Abstract

INTRODUCTIONAlteration of stream hydrology by climate and human sources is predicted to have a major impact on freshwater fishes (Dudgeon et al., ). Stream hydrology is important for the consistency and maintenance of fish habitat, especially for resident or sedentary species, as they are unable to move away from suddenly changing environments. Alterations to stream hydrology, if severe enough, can reduce biodiversity (Pringle, Freeman, & Freeman, ). Determining how stream hydrology affects freshwater fishes is important for determining how to mitigate the impacts of stream hydrology and provide appropriate conservation measures.It is well known that local adaptations to various environments can lead to morphological and genetic differences within the same species (Collin & Fumagalli, ; Endler, ; Kawecki & Ebert, ). Even among similar flow gradients, morphological divergence can occur (McLaughlin & Grant, ; Pakkasmaa & Piironen, ). In freshwater systems, flow has been recognized as a driving force behind intraspecific morphological adaptation in fish species (Brinsmead & Fox, ; Langerhans, ; Langerhans, Layman, Langerhans, & Dewitt, ). In these studies, pelagic fish tend to have more slender, fusiform bodies and deeper caudal peduncles when exposed to faster‐flowing waters (Collin & Fumagalli, ; McLaughlin & Grant, ; Webb, ). These morphological adaptations help minimize drag forces on the body, thereby optimizing energetic expenditure (Sagnes & Statzner, ; Webb, ). Few studies account for benthic fish, however, which have different life histories (Facey & Grossman, ).The Rocky Mountain sculpin (Cottus sp.) is a ‘Threatened’ species in Canada and is listed under the federal Species at Risk Act. Like many legal frameworks around the globe (Vardakas et al., ), the Species at Risk Act provides protection of ‘critical habitat’, defined as ‘the habitat that is necessary for the survival or recovery of a listed wildlife species’ (Fisheries and Oceans Canada, ). In addition, a recovery plan that outlines the actions needed for species recovery is required. This plan must address how to mitigate key threats to the species. For Rocky Mountain sculpin, the main threat is thought to be changes in stream hydrology. Rocky Mountain sculpin live in varying hydrological regimes, differing by three orders of magnitude across its range (from minimum 0–1 m3 s‐1 to >120 m3 s‐1; Figure ). These differences in stream hydrology are due in large part to stream alterations to improve irrigation for agriculture. For example, the St. Mary's River was diverted in northern Montana to the Milk River basin to improve irrigation. In recent years, this has changed the flow in the North Milk River from about 7.5 m3 s‐1 during natural conditions to upwards of 17m3 s‐1 during augmentation. Given the alteration of the natural hydrology of systems occupied by Rocky Mountain sculpin and the difference in magnitude of hydrological regimes, Rocky Mountain sculpin provide a model species for understanding impacts of stream hydrology on differences in population levels for freshwater fishes.1Map of study locations (top) and associated hydrographs (bottom). Fish samples of populations of Rocky Mountain sculpin were collected from the Flathead River (blue), Lee Creek (orange), St. Mary River (red), and the North Milk River (green). Hydrographs were developed from 84 to 104 years of data collected by four representative gauging stations (Water Survey of Canada, )Bottom‐dwelling fish exhibit morphological differences for a variety of reasons, including habitat characteristics such as substrate (Whiteley, Gende, Gharrett, & Tallmon, ), thermal conditions (Koumoundouros et al., ), and flow regime (Natsumeda, Tsuruta, Takeshima, Awata, & Iguchi, ). Sculpins are sedentary fish with relatively deep bodies and a bulky caudal peduncle (Webb, ). Their robust torso allows for short, strong bursts of forward movement when necessary, but prolonged swimming is energetically costly. Sculpins maintain their position in running water by relying heavily on their large, rigid pectoral fins to hold themselves in place among the substrate (Facey & Grossman, ). In highly turbulent environments, sculpin exhibit prominent pectoral fins forming more robust tissue connections where the fin meets the body (Kane & Higham, ).Unlike pelagic species, benthic species must morphologically reduce drag against their body in river systems to maintain their position (Koehl, ). Because Rocky Mountain sculpin have survived in varying hydrological regimes, Rocky Mountain sculpin should have a more streamlined body shape to reduce drag and/or stronger pectoral fins to grip onto substrate in the higher flow systems. Higher velocities may also select for differences in pore and fin ray counts to account for different demands on flow compensation and prey detection rates. This study compared the body morphology and fin ray and pore counts of Rocky Mountain sculpin across four populations in Canada which vary in stream hydrology and biogeography. Specifically, the objectives were to understand: (1) if Rocky Mountain sculpin exhibit morphological adaptations to stream flow across their four prominent river systems in Alberta and British Columbia, Canada, and (2) whether there is a difference in meristic counts between Rocky Mountain sculpin across populations. This information has built upon the known ecology of this species and will help in identifying whether populations from different river basins and biogeography are unique in their meristics and morphometrics.METHODSSpeciesThe identity and description of sculpins (family Cottidae), including Rocky Mountain sculpin, have long confused researchers in western North America (Bajkov, ; Hughes & Peden, ; Lemoine et al., ; Markle & Hill, ; Taylor, ; Zimmerman & Wooten, ). Previously, Rocky Mountain sculpin were confused with at least three other species (COSEWIC, ). Rocky Mountain sculpin were first characterized in 2002 (Neely, ), and are considered a taxonomically valid species in Canada (Taylor, ), despite no official description. Rocky Mountain Sculpin show clear phylogenetic differences with other Cottus species (COSEWIC, ; Neely, ; Young, McKelvey, Pilgram, & Schwartz, ). Differences in population levels do occur between populations on the eastslope and westslope of the Rocky Mountains in Canada (Ruppert et al., ), but analysis of 1140 cytochrome b base pairs show that these populations are the same species (COSEWIC, ). Thus, despite historical confusion over species identity, contemporary data clearly show Rocky Mountain sculpin, from both eastslope and westslope of the Rocky Mountains, as the same, unique species.Study areaRocky Mountain sculpin occur across a wide range of stream hydrology (Figure ). Of the four study systems, the Flathead River has the highest flow, with average peak discharge rate of 125 m3 s‐1. The St. Mary River has the highest average peak flow rate for streams on the eastern slopes of the Rocky Mountains with a mean of 61 m3 s‐1. The next highest flow rate occurs in Lee Creek with average peak flows at 9 m3 s‐1. North Milk River is the slowest river with mean peak flow rate of 8 m3 s‐1. Much of this flow is through augmentation from the St. Mary's River. Built in 1917, the St. Mary canal was designed to divert water from the St. Mary River to the Milk River basin during the growing season to supplement flows for irrigation of crops in the Milk River basin.Sculpin collectionA minimum of 40 Rocky Mountain sculpins were collected from each of the four rivers. Sculpins were electro‐shocked using a Smith‐Root LR‐24 backpack electrofisher. Sculpins were left overnight in a flow‐through bin to reduce their stress levels before transport. In total, 339 live fish were moved to holding tanks in the Aquatics Research Facility at the University of Alberta, a level‐three bio‐secure aquatic holding facility.Sculpin data preparationRocky Mountain sculpin collected from each river were euthanized and positioned with splayed fins. Images were taken with a Nikon D3100 digital single‐lens reflex camera equipped with a Nikon DX AF‐S Nikkor 18–55 mm zoom lens set at 35 mm, 225 ISO, 5.3 f‐stop, and 1/60 shutter speed. Torsos were placed at a level plane with the camera lens from head to caudal fin to prevent a warped, disproportionate representation of shape in the digital photos. Digital photos of the dorsal and lateral perspectives of the fish were taken in RAW (.NREF) form. Each photograph included a reference scaling factor to standardize fish size across photographs. Meristic measurements, such as fin rays, fin spines, and head/body pores, were subsequently determined for each individual. These meristic counts were chosen as they were similar to the original species description (2002).Landmarks were placed in areas that provided easy replication, such as the location of the eye and the origin and insertion of fin locations. Landmarks were marked using tpsDIG software (Rohlf, ), and their location was translated into X and Y coordinates in a .TPS file. Landmark criteria, as described by Dryden and Mardia (), were reference points on the sculpin's body that could be found accurately and marked across a large number of specimens. These points included fin insertion points, eye placement, and caudal peduncle locations.Differences in body morphologyDifferences in body morphology were tested using geometric morphometric analysis using the geomorph in the R programming language (Adams, Collyer, & Sherratt, ; R Core Team, ). A General Procrustes Analysis (GPA) was conducted on the coordinates. The GPA optimally superimposed landmarks by rotating, sizing and centralizing them without compromising the overall shape from the coordinates (Rohlf, ; Rohlf & Marcus, ; Slice, ). This process produced useable X‐Y Procrustes residual coordinates that could be used for a variety of multivariate analyses. In addition, the difference between Euclidean distances between the Procrustes landmarks was measured and compared for each population. Euclidean Distance Matrix Analysis (EDMA) was used to identify variance across the mean landmark positions between groups (Lele, ).To determine if there were differences in overall shape across populations, a distance‐based Procrustes Analysis of Variance (ANOVA) using 1000 permutations was used to test between GPA residuals of dorsal and lateral landmarks. The Procrustes ANOVA allowed a simultaneous analysis of sum‐of‐squared Procrustes distances across all the coordinates instead of considering one coordinate at a time (Goodall, ). This allowed the determination of significant differences in overall shape across populations using permutation analyses to generate reliable P‐values (Goodall, ; Sherratt, ). Pairwise comparisons were conducted using permutational t‐tests, with 1000 resampling iterations to determine which populations were significantly different using pairwiseD.test in geomorph. This test is specialized for comparing geometric shape variation across groups by using the Procrustes residual values that represent shape variation, and calculating P‐values based on the Euclidean distances between the mean of each population (Adams et al., ; Collyer, Sekora, & Adams, ). To visualize the results of the pairwise comparisons, coordinate data for each sculpin were converted into a single warp score using candisc (Friendly & Fox, ). Warp scores were generated based on the degree and location of morphometric variance relative to the other fish in the study (Webster & Sheets, ; Zelditch, Swiderski, & Sheets, ). Means of the values in each population were calculated as well as the Euclidean distance between those means (Sherratt, ). After 1000 permutations, a Canonical Variates Analysis (CVA) was used to determine differences between groups that were calculated in the pairwise comparison. As a form of discriminatory analysis, CVA was used because it can measure differences in overall shape between the four predefined vectors (Rohlf, ; Zelditch et al., ). The vectors identified landmarks that were major drivers in population variation. This generated the data necessary for visualization and the vegan package (Oksanen et al., ) was used to help customize the visualization.Meristic differences in body formFin ray and pore counts were compared across populations using a permutational ANOVA. Permutational ANOVA is useful when datasets do not adhere to the assumptions of a traditional ANOVA (i.e. independence, normality, and homogeneity of variances), which is common in natural and biological realms (Anderson, ). To test significance across groups, a pairwise Wilcoxon Rank Sum Test was used and a Holm P‐value adjustment was applied to correct for Type I errors when multiple comparisons are made (Aickin & Gensler, ).RESULTSDifferences in body morphologyThere was strong divergence in body morphology of Rocky Mountain sculpin across its Canadian range. The Flathead River population was the most visibly divergent population among the four populations of Rocky Mountain sculpin. From the dorsal perspective, the caudal peduncle and snout positions relative to pectoral fin insertion points were the primary features that differentiated the westslope population from the eastslope groups (P<0.001; Figures  and ). In addition, westslope Rocky Mountain sculpins had stunted heads, creating a more antero‐posteriorly compressed head shape, and the length of the torso beyond the pectoral fin insertion points was significantly narrower and longer. From the lateral perspective, the westslope population was significantly different from the eastslope populations (P<0.001; Table ). Eye placement was oriented closer to the insertion point of the dorsal spines, contributing to a more dorsally flattened form (Figure ). The placement of the isthmus showed significant differentiation in the Flathead River population (Figure , P < 0.01). It was located closer to the snout, probably as the result of a more antero‐posteriorly compressed head, as seen in the dorsal view. The westslope population also had the longest, narrowest, and flattest body shape of the four groups.2Differences in dorsal Procrustes scores of Rocky Mountain sculpin (Cottus sp.) grouped by river system. Using Canonical Variate analysis, the distance between centroids (crosshairs) of populations was maximized. Populations included the Flathead River (blue), Lee Creek (orange), St. Mary River (red), and North Milk River (green)3Differences in lateral Procrustes scores of Rocky Mountain sculpin (Cottus sp.) grouped by river system. Using Canonical Variate analysis, the distance between centroids (crosshairs) of populations was maximized. Populations included the Flathead River (blue), Lee Creek (orange), St. Mary River (red), and North Milk River (green)4Dorsal (top) and lateral (bottom) landmarks of Rocky Mountain sculpin (Cottus sp.) across its Canadian range, including the Flathead River (blue), Lee Creek (orange), St. Mary River (red), and North Milk River (green). Dorsal landmark placements were amplified 4.5 times to show the differences across populations. Lateral landmark placements had to be amplified by only three times to show differences visually. The X and Y axes represent the vector scores once standardized using General Procrustes AnalysisPairwise differences in morphometric landmarks of Rocky Mountain sculpin (Cottus sp.) across populations including the Flathead River, Lee Creek, North Milk River, and St. Mary River. Shown are permutated t‐tests using Euclidean distances between the means of dorsal (top corner) and lateral (bottom corner) landmarks across each river. P‐values are based on 1000 permutations and are adjusted using a Holm correction. Asterisks indicate significant differencesFlatheadLee CreekNorth MilkSt. MaryFlathead‐0.001*0.001*0.001*Lee Creek0.001*‐0.001*0.13North Milk0.001*0.001*‐0.001*St. Mary0.001*0.001*0.001*‐Differences in body morphology were strongly divergent not just across the continental divide, but also within the eastslope populations. All of the three eastslope populations showed a significant difference in shape (P <0.001; Table ). The differences in the North Milk River population were pronounced in the pectoral fin insertion point locations (Figure ). The width between the anterior insertion points in the dorsal view showed that the North Milk River population had the widest overall shape (Figure ). The close proximity of the posterior and anterior pectoral fin insertion points led to the North Milk River population having the smallest fin base (Figure ). Lee Creek sculpins had lower and posteriorly placed pelvic fin insertion points, leading to the most dorso‐ventrally broad shape of the four populations (Figure ). St. Mary River sculpin eye placement was closest to the snout and the dorsal spine insertion point is almost level with the eyes, contributing to the most dorso‐ventrally flattened eastslope population (Figure ).Meristic differences in body formThere were significant meristic differences in body form of Rocky Mountain sculpin across its Canadian range. The average pore counts and fin ray/spine counts among populations were significantly different (permANOVA, P<0.001). Pore counts varied across populations by no more than five pores and fin ray/spine counts varied only by one in populations that were significantly different (Table ).Summary of meristic fin and pore counts of Rocky Mountain sculpin (Cottus sp.) across its Canadian range, including the Flathead River, Lee Creek, the North Milk River, and the St. Mary River. Shown are means, ranges, and standard deviationsRiver PopulationsFlatheadLee CreekNorth MilkSt. MaryCharacterMeanRangeSDMeanRangeSDMeanRangeSDMeanRangeSDPoresInfraorbital106‐10±1.5108‐10±0.3108‐14±1.1106‐10±0.5Pre‐operculo‐mandibular1410‐16±1.31411‐18±0.91412‐18±1.11412‐16±0.6Mandibular1310‐16±1.41712‐18±1.5157‐20±2.91710‐18±1.6Posterior orbital1712‐22±2.11611‐18±1.71711‐22±2.41710‐22±2.6Forehead22±011‐3±0.421‐2±0.511‐2±0.4Lateral line2419‐32±2.72415‐30±3.12611‐34±4.22314‐36±3.0Fin ray/spine countsDorsal spines86‐9±0.787‐9±0.687‐9±0.687‐8±0.5Dorsal rays1713‐19±1.21716‐19±0.91715‐20±0.81713‐19±1.3Anal rays1211‐14±0.81211‐13±0.71211‐15±0.71210‐13±0.8Pectoral rays1311‐14±0.61210‐13±1.1129‐13±1.0129‐14±1.0Pelvic rays44±044±044±044±0Caudal rays1312‐14±0.51210‐13±0.8129‐13±0.91211‐15±1.0Opercular spines11±011±011±011±0N4174106118The westslope population exhibited several significant meristic differences from the eastslope populations. They had significantly more forehead pores (P<0.001), and fewer mandibular pores (P<0.001; Tables  and ). Furthermore, the westslope population had more pectoral (P<0.001) and caudal fin rays (P<0.001) than the Lee Creek and North Milk River population (Tables  and ).Pairwise differences in pore counts of Rocky Mountain sculpin (Cottus sp.) across Canada, including the Flathead River, Lee Creek, the North Milk River, and the St. Mary River. Pairwise Wilcoxon Rank Sum Test P‐values are displayed with Holm corrected significance levels. Asterisks indicate significant resultsInfraorbital poreMandibular poreForehead poreLateral line poreFlathead : Lee Creek0.91<0.001*<0.001*0.67Flathead : North Milk1.00<0.001*<0.001*0.02Flathead : St. Mary0.81<0.001*<0.001*0.06Lee Creek : North Milk0.220.002*<0.001*0.01Lee Creek : St. Mary1.000.481.000.21North Milk : St. Mary0.080.002*<0.001*<0.001*Pairwise differences in fin ray and spine counts of Rocky Mountain sculpin (Cottus sp.) across Canada, including the Flathead River, Lee Creek, the North Milk River, and the St. Mary River. Pairwise Wilcoxon Rank Sum Test P‐values are displayed with Holm corrected significance levels. Asterisks indicate significant resultsDorsal fin spinesDorsal fin raysAnal fin raysPectoral fin raysCaudal fin raysFlathead : Lee Creek0.640.991.00<0.001*<0.001**Flathead : North Milk0.071.001.00<0.001*<0.001**Flathead : St. Mary0.051.001.00<0.001*0.02Lee Creek : North Milk0.730.991.001.000.81Lee Creek : St. Mary0.731.001.001.000.81North Milk : St. Mary0.731.001.000.710.81There were some meristic variations among the eastslope populations. Lee Creek and North Milk River populations differed in three of the pore counts, where the Lee Creek population had significantly more mandibular pores (P=0.002), but fewer forehead (P<0.001) pores (Tables ). Similar to Lee Creek, the St. Mary River population had a higher number of mandibular pores (P=0.002) than the North Milk River population, as well as fewer forehead (P<0.001) and lateral line pores (P<0.001; Tables  and ). There were no differences in fin ray/spine counts across east slope populations. There were no meristic differences between the Lee Creek and St. Mary populations.DISCUSSIONMany species exhibit morphological divergence in relation to environmental gradients (Endler, ; Kawecki & Ebert, ). Rocky Mountain sculpin exhibited strong morphological and meristic body differences across broad hydrological gradients. Specifically, higher flow rates were correlated with higher adaptive morphological divergence. For example, populations exhibited strong divergence in meristics across the hydrological gradient. The population with highest flow (westslope) had more caudal fin rays, which help provide thrust, and pectoral fin rays to increase friction against substrate in high flow environments (Kane & Higham, ; Taft, Lauder, & Madden, ; Webb, ). The highest flow population also had fewer mandibular pores and more forehead pores than the lower flow (eastslope) population. In lotic systems, sculpin use their head pores to detect suspended prey and mandibular pores to extract buried prey (Hoekstra & Janssen, ). The increase in forehead pores helps to detect drifting prey in fast currents. The reduction in mandibular pores is an example of a feature that cannot be justified by flow regime, and is possibly the result of divergence based on prolonged biogeographic isolation from other populations.Biogeographic distance can play an important role in determining morphological divergence. The continental divide has undoubtedly influenced body morphology between westslope and eastslope Rocky Mountain sculpin populations given that the populations have probably been isolated for about 10 000 years (Nelson & Paetz, ). The westslope populations exhibited a more compressed and elongated torso than the eastslope populations, probably because of the interplay between prolonged biogeographic isolation and different flow regimes (Langerhans et al., ). If flow regime were the primary cause of variation, there would have been more similar morphologies between the St. Mary and Flathead River populations, and almost identical morphologies between the Lee Creek and North Milk River populations. Instead, there were closer morphometric and meristic values between the Lee Creek and St. Mary River populations (the closest geographically), as Lee Creek is a tributary of the St. Mary River. This theory is also validated by the large degree of phenotypic variation between the westslope populations and the eastslope populations. Gene flow is a major influence on the success of local adaptation (Kawecki & Ebert, ), therefore the disconnection from the eastslope populations undoubtedly contributes to the observed morphological divergence.Phenotypic variation is required to optimize fitness, especially in relation to stochastic environmental events (McGuigan, Franklin, Moritz, Blows, & Wainwright, ; Taylor & McPhail, ). Since phenotypic diversity parallels genetic diversity in Rocky Mountain sculpin (Ruppert et al., ), there is the possibility that under persistent selection Rocky Mountain sculpin will have maladaptive morphological features with changing stream flows. For example, under baseflow conditions, Rocky Mountain sculpin with narrower and shorter caudal peduncles were shown to have poorer swimming performance (Veillard, Ruppert, Tierney, Watkinson, & Poesch, ). With climate change expected to rapidly alter stream hydrology, these traits may be maladaptive, making it difficult for Rocky Mountain sculpin to adapt to rapidly changing conditions (Stearns & Kawecki, ).Understanding how biogeography and stream hydrology influences body morphology is important to improve species conservation. Rocky Mountain sculpin are a threatened species in Canada and could be further imperilled if their dominant phenotypes are incapable of adapting to altered river conditions such as flood events and drought conditions (Lytle & Poff, ). Their limited dispersal can lead to specialized morphologies and extreme events could eliminate a population. Management efforts should be directed, therefore, toward preserving genetic diversity at the population level of the species (Riffel & Schreiber, ; Ruppert et al., ), while focusing future research toward understanding how genotypic and phenotypic divergence change in relation to stream hydrology. Almost all fisheries management programmes are developed around preserving genetic variation, thereby protecting the species’ ability to overcome unpredicted environmental circumstances (Fraser & Bernatchez, ).CONCLUSIONMany studies have shown that life‐history characteristics are good predictors of extinction risk. These studies provide an important linkage to how biological features such as reproductive traits, habitat, and age/growth relationships can make species more susceptible to decline or put them more generally at risk (Glass, Corkum, & Mandrak, ; Purvis, Gittleman, Cowlishaw, & Mace, ; Stark, Banks, & Vargas, ). These relationships have intuitive appeal as it is easy to see the linkages between life‐history characteristics such as small body size, and understand how they may lead to limited dispersal and potentially a lack of connectivity between populations (and potential rescue). However, despite the importance of phenotypic plasticity in shaping local adaptations, the utility of phenotypic or morphological diversity has remained relatively understudied in the conservation literature. Here, a combination of biogeography and flow regime appear to be driving phenotypic and morphological divergence between Rocky Mountain sculpin populations. High‐flow lotic systems have shifted Rocky Mountain sculpin towards minimizing body depth. Results from this study suggest that small physical differences within the range of a fish species can have impacts on the energetic ability of the Rocky Mountain sculpin to exist in a wide range of lotic environments. By incorporating morphological diversity within the context of life‐history considerations already being used in assessing species, the potential of species to respond to both natural and human alterations can be predicted more accurately.ACKNOWLEDGEMENTSFunding for this project was provided by Fisheries and Oceans Canada, Species at Risk (SARCEP) (DW, MP), and NSERC Discovery Grant (MP). The authors thank Marie Veillard, Denyse Dawe, Christopher Smith, Elashia Young, Caitlin Good, Kenton Neufeld, Bryan Maitland, and Doug Watkinson for assisting with field collection and laboratory work. The care and assistance importing fish to the University of Alberta Aquatics Facility was provided by Toni Bayans and Simmone Kerswell. Additional thanks to Dr Jonathan Ruppert for his helpful advice on data analysis and Hedin Nelson‐Chorney for his critiques on early manuscript drafts. The procedures for this project was conducted under an approved Canadian Council of Animal Care Approval (AUP 00000759), Alberta Fish Research Licences 14‐2415AFR and 16‐0101FR, Alberta Fish Import Licence FIL2015‐0074, Federal Introductions and Transfers Licence VI16‐236322, and British Columbia Fish Collection Permit CB15‐171090.REFERENCESAdams, D., Collyer, M., & Sherratt, E. (2014). geomorph: Software for geometric morphometric analyses. R package version, 2.Aickin, M., & Gensler, H. (1996). Adjusting for multiple testing when reporting research results: The Bonferroni vs Holm methods. American Journal of Public Health, 86, 726–728.Anderson, M. J. (2001). Permutation tests for univariate or multivariate analysis of variance and regression. Canadian Journal of Fisheries and Aquatic Sciences, 58, 626–639.Bajkov, A. (1927). Reports of the Jasper Park Lakes Investigations, 1925‐26. I. The fishes. Contributions to Canadian Biology and Fisheries, 3, 379–404.Brinsmead, J., & Fox, M. (2002). Morphological variation between lake‐ and stream‐dwelling rock bass and pumpkinseed populations. Journal of Fish Biology, 61, 1619–1638.Collin, H., & Fumagalli, L. (2011). Evidence for morphological and adaptive genetic divergence between lake and stream habitats in European minnows (Phoxinus phoxinus, Cyprinidae). Molecular Ecology, 20, 4490–4502.Collyer, M., Sekora, D., & Adams, D. (2015). A method for analysis of phenotypic change for phenotypes described by high‐dimensional data. Heredity, 115, 357–365.COSEWIC. (2010). COSEWIC assessment and status report on the Rocky Mountain Sculpin Cottus sp., Westslope populations, in Canada. Committee on the Status of Endangered Wildlife in Canada. Ottawa.Dryden, I. L., & Mardia, K. V. (2016). Statistical shape analysis: With applications in R. Chichester, UK: John Wiley & Sons.Dudgeon, D., Arthington, A. H., Gessner, M. O., Kawabata, Z. I., Knowler, D. J., Lévêque, C., … Stiassny, M. L. (2006). Freshwater biodiversity: Importance, threats, status and conservation challenges. Biological Reviews, 81, 163–182.Endler, J. A. (1986). Natural selection in the wild. Princeton, NJ: Princeton University Press.Facey, D. E., & Grossman, G. D. (1990). The metabolic cost of maintaining position for four North American stream fishes: Effects of season and velocity. Physiological Zoology, 63, 757–776.Fisheries and Oceans Canada. (2017). Action Plan for the Milk River and St. Mary River Drainage Basins in Canada. Fisheries and Oceans, Species at Risk Act Action Plan Series, Ottawa, Canada.Fraser, D. J., & Bernatchez, L. (2001). Adaptive evolutionary conservation: Towards a unified concept for defining conservation units. Molecular Ecology, 10, 2741–2752.Friendly, M., & Fox, J. (2010). Candisc: R package for canonical discriminant analysis: Accessed at: http://cran.r‐project.org/web/packages/candisc.Glass, W. R., Corkum, L. D., & Mandrak, N. E. (2017). Living on the edge: Traits of freshwater fish species at risk in Canada. Aquatic Conservation: Marine and Freshwater Ecosystems, 27, 938–945.Goodall, C. (1991). Procrustes methods in the statistical analysis of shape. Journal of the Royal Statistical Society. Series B (Methodological), (1991), 285–339.Hoekstra, D., & Janssen, J. (1985). Non‐visual feeding behavior of the mottled sculpin, Cottus bairdi, in Lake Michigan. Environmental Biology of Fishes, 12, 111–117.Hughes, G. W., & Peden, A. E. (1984). Life history and status of the shorthead sculpin (Cottus confusus: Pisces, Cottidae) in Canada and the sympatric relationship to the slimy sculpin (Cottus cognatus). Canadian Journal of Zoology, 62, 306–311.Kane, E. A., & Higham, T. E. (2012). Life in the flow lane: Differences in pectoral fin morphology suggest transitions in station‐holding demand across species of marine sculpin. Zoology, 115, 223–232.Kawecki, T. J., & Ebert, D. (2004). Conceptual issues in local adaptation. Ecology Letters, 7, 1225–1241.Koehl, M. (1984). How do benthic organisms withstand moving water? American Zoologist, 24, 57–70.Koumoundouros, G., Ashton, C., Sfakianakis, D., Divanach, P., Kentouri, M., Anthwal, N., & Stickland, N. (2009). Thermally induced phenotypic plasticity of swimming performance in European sea bass Dicentrarchus labrax juveniles. Journal of Fish Biology, 74, 1309–1322.Langerhans, R. B. (2008). Predictability of phenotypic differentiation across flow regimes in fishes. Integrative and Comparative Biology, 48, 750–768.Langerhans, R. B., Layman, C. A., Langerhans, A. K., & Dewitt, T. J. (2003). Habitat‐associated morphological divergence in two Neotropical fish species. Biological Journal of the Linnean Society, 80, 689–698.Lele, S. (1993). Euclidean distance matrix analysis (EDMA): Estimation of mean form and mean form difference. Mathematical Geology, 25, 573–602.Lemoine, M., Young, M. K., McKelvey, K. S., Eby, L., Pilgram, K. L., & Schwartz, M. K. (2014). Cottus schitsuumsh, a new species of sculpin (Scorpaeniformes: Cottidae) in the Columbia River basin, Idaho‐Montana, USA. Zootaxa, 3755, 241–258.Lytle, D. A., & Poff, N. L. (2004). Adaptation to natural flow regimes. Trends in Ecology & Evolution, 19, 94–100.Markle, D. F., & Hill, D. L. (2000). Taxonomy and distribution of the Malheur mottled sculpin, Cottus bendirei. Northwest Science, 74, 202–211.McGuigan, K., Franklin, C. E., Moritz, C., Blows, M. W., & Wainwright, P. (2003). Adaptation of rainbow fish to lake and stream habitats. Evolution, 57, 104–118.McLaughlin, R. L., & Grant, J. W. (1994). Morphological and behavioural differences among recently‐emerged brook charr, Salvelinus fontinalis, foraging in slow‐ vs. fast‐running water. Environmental Biology of Fishes, 39, 289–300.Natsumeda, T., Tsuruta, T., Takeshima, H., Awata, S., & Iguchi, K. I. (2014). Variation in morphological characteristics of Japanese fluvial sculpin related to different environmental conditions in a single river system in eastern Japan. Ecology of Freshwater Fish, 23, 114–120.Neely, D. A. (2002). A systematic and taxonomic reassessment of the mottled sculpin species complex, Cottus bairdii Girard (Teleostei: Cottidae) (PhD thesis). University of Alabama, Tuscaloosa, AB.Nelson, J. S., & Paetz, M. J. (1992). The fishes of Alberta. Edmonton, Alberta: University of Alberta Press.Oksanen, J., Blanchet, F., Kindt, R., Legendre, P., Minchin, P., O’Hara, R., … Stevens, H. (2013). Vegan: Community Ecology Package. 2013. R‐package version 2.0‐10. Available from: https://cran.r‐project.org/package=vegan.Pakkasmaa, S., & Piironen, J. (2000). Water velocity shapes juvenile salmonids. Evolutionary Ecology, 14, 721–730.Pringle, C. M., Freeman, M. C., & Freeman, B. J. (2000). Regional effects of hydrologic alterations on riverine macrobiota in the New World: Tropical–temperate comparisons. BioScience, 50, 807–823.Purvis, A., Gittleman, J. L., Cowlishaw, G., & Mace, G. M. (2000). Predicting extinction risk in declining species. Proceedings of the Royal Society of London B: Biological Sciences, 267, 1947–1952.R Core Team (2016). R: A language and environment for statistical computing. Vienna, Austria: R Foundation for Statistical Computing.Riffel, M., & Schreiber, A. (1998). Morphometric differentiation in populations of the Central European sculpin Cottus gobio L., a fish with deeply divergent genetic lineages. Canadian Journal of Zoology, 76, 876–885.Rohlf, F. J. (1999). Shape statistics: Procrustes superimpositions and tangent spaces. Journal of Classification, 16, 197–223.Rohlf, F. J. (2005). tpsDig, digitize landmarks and outlines, version 2.05. Department of Ecology and Evolution, State University of New York at Stony Brook.Rohlf, F. J., & Marcus, L. F. (1993). A revolution in morphometrics. Trends in Ecology & Evolution, 8, 129–132.Ruppert, J. L., James, P. M., Taylor, E. B., Rudolfsen, T., Veillard, M., Davis, C. S., … Poesch, M. S. (2017). Riverscape genetic structure of a threatened and dispersal limited freshwater species, the Rocky Mountain Sculpin (Cottus sp.). Conservation Genetics, 2017, 1–13.Sagnes, P., & Statzner, B. (2009). Hydrodynamic abilities of riverine fish: A functional link between morphology and velocity use. Aquatic Living Resources, 22, 79–91.Sherratt, E. (2014). Quick guide to Geomorph v. 2.0. Available from: http://www.public.iastate.edu.Slice, D. E. (2001). Landmark coordinates aligned by Procrustes analysis do not lie in Kendall's shape space. Systematic Biology, 50, 141–149.Stark, J. D., Banks, J. E., & Vargas, R. (2004). How risky is risk assessment: The role that life history strategies play in susceptibility of species to stress. Proceedings of the National Academy of Sciences of the United States of America, 101, 732–736.Stearns, S. C., & Kawecki, T. J. (1994). Fitness sensitivity and the canalization of life‐history traits. Evolution, (48), 1438–1450.Taft, N., Lauder, G., & Madden, P. (2008). Functional regionalization of the pectoral fin of the benthic longhorn sculpin during station holding and swimming. Journal of Zoology, 276, 159–167.Taylor, E. B. (2010). Changes in taxonomy and species distributions and their influence on estimates of faunal homogenization and differentiation in freshwater fishes. Diversity and Distributions, 16, 676–689.Taylor, E. B., & McPhail, J. (1985). Variation in body morphology among British Columbia populations of coho salmon, Oncorhynchus kisutch. Canadian Journal of Fisheries and Aquatic Sciences, 42, 2020–2028.Vardakas, L., Kalogianni, E., Papadaki, C., Vavalidis, T., Mentzafou, A., & Koutsoubas, D. (2017). Defining critical habitat conditions for the conservation of three endemic and endangered cyprinids in a Mediterranean intermittent river before the onset of drought. Aquatic Conservation: Marine and Freshwater Ecosystems. https://doi.org/10.1002/aqc.2735.Veillard, M. F., Ruppert, J. L., Tierney, K., Watkinson, D. A., & Poesch, M. (2017). Comparative swimming and station‐holding ability of the threatened Rocky Mountain Sculpin (Cottus sp.) from four hydrologically distinct rivers. Conservation Physiology, 5, 1–12.Water Survey of Canada. (2016). Wateroffice: Historical Hydrometric Data: Environment Canada, Government of Canada.Webb, P. W. (1984). Form and function in fish swimming. Scientific American, 251, 58–68.Webster, M., & Sheets, H. D. (2010). A practical introduction to landmark‐based geometric morphometrics. Quantitative Methods in Paleobiology, 16, 168–188.Whiteley, A. R., Gende, S. M., Gharrett, A. J., & Tallmon, D. A. (2009). Background matching and color‐change plasticity in colonizing freshwater sculpin populations following rapid deglaciation. Evolution, 63, 1519–1529.Young, M. K., McKelvey, K. S., Pilgram, K. L., & Schwartz, M. K. (2013). DNA barcoding at riverscape scales: Assessing biodiversity among fishes of the genus Cottus (Teleostei) in northern Rocky Mountain streams. Molecular Ecology Resources, 13, 583–595.Zelditch, M. L., Swiderski, D. L., & Sheets, H. D. (2012). Geometric morphometrics for biologists: A primer (2nd ed.). San Diego, CA: Elsevier.Zimmerman, E. G., & Wooten, M. C. (1981). Allozymic variation and natural hybridization in sculpins, Cottus confusus and Cottus cognatus. Biochemical Systematics and Ecology, 9, 341–346.

Journal

Aquatic Conservation: Marine and Freshwater Ecosystems – Wiley

Published: Jan 1, 2018

Keywords: ; ; ;

There are no references for this article.